Navigation method based on three-dimensional cubic space gridding algorithm

Abstract

This invention relates to the field of low-altitude unmanned aerial vehicle (UAV) navigation and airspace control technology, specifically disclosing a navigation method based on a three-dimensional spatial gridding algorithm. This navigation method includes the following steps: S1, Airspace gridding: Dividing the 0-600m controlled low-altitude airspace into several 10m side-length regular hexagonal prism honeycomb three-dimensional grids, each grid is uniquely identified using Grid(X,Y,Z) three-dimensional coding. This invention abandons the traditional navigation model that relies solely on planar latitude and longitude for route planning. It uses a 10m side-length regular hexagonal prism honeycomb three-dimensional grid to achieve fine segmentation of the entire 0-600m low-altitude airspace. The X / Y / Z axes simultaneously constrain the geographical location and flight altitude of navigation points, allowing for spatial differentiation of different aircraft flight levels during the route planning stage.

Description

Technical Field

[0001] This invention relates to the field of low-altitude unmanned aerial vehicle navigation and airspace control technology, and in particular to a navigation method based on a three-dimensional spatial gridding algorithm. Background Technology

[0002] Currently, most low-altitude unmanned aerial vehicles (UAVs) on the market adopt the traditional two-dimensional planar coordinate navigation mode. Navigation path planning relies solely on ground latitude and longitude planar point calculations, completely ignoring the vertical spatial attributes of the 600-meter low-altitude airspace. This is a core shortcoming of the existing low-altitude navigation system. When designing flight routes, existing navigation schemes rely solely on the aircraft's onboard navigation module to autonomously detect surrounding obstacles and the positions of nearby aircraft in real time. The onboard navigation system cannot access unified airspace control data across the entire region, lacks standardized three-dimensional airspace spatial division benchmarks, and cannot incorporate refined flight route constraints based on the legal safety intervals for different aircraft types. Due to this navigation logic limitation, when UAVs with different takeoff weights and different safety distance standards operate in the same low-altitude airspace, the flight routes autonomously planned by the onboard navigation are prone to overlap at the vertical altitude level. Once multiple aircraft converge at close range, the onboard navigation can only rely on its own radar for temporary emergency avoidance, lacking a pre-allocation of airspace and pre-planning of flight routes. Meanwhile, existing navigation systems lack the capability to calculate airspace grid capacity, making it impossible to predict the remaining capacity of the airspace at target flight points in advance. When the target flight airspace becomes saturated and congested, aircraft navigation can only temporarily and randomly change routes, which significantly extends the operational range, increases aircraft power consumption, and is highly susceptible to close-range flight conflicts caused by temporary route changes. In addition, existing navigation systems mostly operate on a mode where aircraft actively report their positions and autonomously decide their routes. The simultaneous reporting of position data by a large number of aircraft can cause data congestion and delays in command issuance on the airspace control platform, further exacerbating the risk of navigation loss of control and severely restricting the large-scale and routine operation of UAVs in low-altitude airspace below 600 meters. Therefore, to solve these problems, a navigation method based on a three-dimensional spatial gridding algorithm is proposed. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by proposing a navigation method based on a three-dimensional spatial gridding algorithm.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A navigation method based on a three-dimensional spatial meshing algorithm, comprising the following steps:

[0006] S1. Airspace Gridding: The controlled low-altitude airspace from 0 to 600m is divided into several 10m side-length regular hexagonal prism honeycomb three-dimensional grids. Each grid is uniquely identified by a Grid(X,Y,Z) three-dimensional code. The Z-axis number 0 corresponds to a height of 0-10m, and the maximum Z value of 59 corresponds to a height of 590-600m. A dynamic correction coefficient K is configured for each grid according to its geographical location attributes: K=0.6 for urban core areas / airport airspace, K=0.8 for suburbs / non-sensitive areas, and K=1.0 for remote uninhabited areas.

[0007] S2, Model Parameter Pre-storage: Pre-enter the legal safety interval S corresponding to the five types of drones: micro, light, small, medium and large, and calculate the required safe space volume of a single model by using Voccupied=(2S)³;

[0008] S3. Grid capacity calculation: The theoretical number of flights can be obtained by rounding down N = (single grid volume ÷ Voccupied), and the actual capacity Nreal = rounded down (N × K) can be calculated by combining the correction factor K.

[0009] S4. Point-to-point grid positioning: Based on the latitude, longitude, and flight altitude of the UAV navigation starting point and target point, the corresponding three-dimensional grid code is calculated.

[0010] S5. Navigation Path Allocation: Micro UAVs use individual idle grids connected in series to form navigation routes. Light, small, medium and large UAVs calculate the cluster side length according to the safety interval. A single navigation node is formed by continuous grid clusters, and multiple clusters are spliced ​​together to form a complete navigation path.

[0011] S6. Passive navigation command issuance: After the control platform locks the grid resources corresponding to the route, it issues navigation commands to the UAV's onboard navigation terminal. The UAV passively receives the commands and completes navigation flight along the specified grid path.

[0012] S7. Grid Resource Release: After the drone leaves the corresponding grid airspace, the control platform releases the grid resources and updates the number of times the grid has been occupied.

[0013] Preferably, in step S1, the grid code conversion method is as follows: the longitude value is multiplied by a preset conversion factor to obtain the X number, the latitude value is multiplied by a preset conversion factor to obtain the Y number, and the altitude value is divided by 10 and rounded down to obtain the Z number. The control platform stores the codes, area types, no-fly status, and occupied flight data of all grids in key-value pairs.

[0014] Preferably, in step S2, the legal parameters for the five types of drones are configured as follows: 5m safety interval for micro drones, 10m safety interval for light drones, 15m safety interval for small drones, 20m safety interval for medium drones, and 30m safety interval for large drones; micro drones have a single grid Nreal ≥ 1, while light, small, medium, and large drones have a single grid Nreal = 0, and cannot occupy a single grid to complete navigation deployment.

[0015] Preferably, in step S5, the formula for calculating the side length of the grid cluster is: =Rounded up (2S / 10), the cluster is × When assigning a cluster to a continuous cube grid, the system checks the existence and no-fly zones of all grids within the cluster sequentially. If any grid is missing or is no-fly, the cluster assignment is considered a failure.

[0016] Preferably, in step S6, the navigation instructions issued by the control platform are divided into three categories: access navigation instructions, capacity reminder instructions, and avoidance and rerouting instructions.

[0017] The access navigation instructions include the route grid / grid cluster code, flight altitude range, and maximum speed limit parameters for the aircraft type.

[0018] The capacity indicator displays the remaining available aircraft sorties for the target grid.

[0019] The evasion rerouting command is used when the target grid is fully loaded or no-fly zone, and includes recommended available alternative grid coordinates.

[0020] Preferably, the method for selecting alternative grids to avoid rerouting instructions is as follows: search the target grid's six adjacent grids along the X / Y / Z axes, and select free grids that are not no-fly zones and have been occupied by fewer than Nreal as alternative navigation points.

[0021] Preferably, it also includes a temporary no-fly control step. The control platform can set any one or more grids to a temporary no-fly state as needed. The grid marked as no-fly will no longer participate in the navigation route allocation of any aircraft type. If the drone navigation target falls on the no-fly grid, the avoidance and rerouting logic will be triggered directly.

[0022] Preferably, the airborne navigation terminal does not have autonomous route planning or full-domain airspace detection and calculation functions. It is only equipped with an instruction parsing module and a route following control module. The UAV executes navigation flight according to the instructions issued by the control platform throughout the entire process, without needing to transmit its own position data back to the control platform in real time.

[0023] Preferably, it also includes a congestion verification step: after the drone initiates a navigation request, the platform first calculates the target grid Nreal and the currently occupied flights Nused. When Nused < Nreal, it determines that access is allowed and locks the grid resources; when Nused ≥ Nreal, it determines that the grid is full and directly outputs an avoidance and rerouting command.

[0024] Preferably, in step S7, after the UAV completes grid interval navigation and leaves the field, the onboard terminal sends a short departure message to the control platform. After receiving the message, the control platform automatically decrements the occupied count value of the corresponding grid by 1. The released free grid can be used for navigation resource allocation for subsequent new UAVs.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] 1. This invention abandons the traditional navigation model that relies solely on planar latitude and longitude for route planning. It uses a 10m side-length regular hexagonal prism honeycomb three-dimensional grid to achieve fine segmentation of the entire low-altitude airspace from 0 to 600 meters. The X / Y / Z axes simultaneously constrain the geographical location and flight altitude of navigation points. During the route planning stage, different aircraft flight levels can be distinguished from spatial dimensions. Compared with traditional two-dimensional navigation, which cannot control vertical flight spacing and is prone to problems such as close-range intersecting flights at different altitudes with the same latitude and longitude, this solution uses an independent grid as the smallest spatial unit when laying out navigation lines. The spatial position of the navigation route of any aircraft type is accurately anchored by the grid coordinates. Combined with the safe volume occupied by the aircraft type, the spatial boundary is defined. From the source of navigation, spatial overlap of different aircraft routes in any longitudinal, lateral, or vertical dimension is avoided, which greatly reduces the probability of low-altitude flight collisions and improves the stability of UAV navigation operations in the entire airspace.

[0027] 2. This invention calculates the number of aircraft each grid can support in real time based on the aircraft type safety interval and the area correction coefficient. At the beginning of the UAV's navigation application, it can check the remaining available resources of the grid through which the target route passes. If the grid is full, it directly pushes the route to the surrounding alternative grids. Compared with the existing navigation method of waiting for the aircraft to arrive in the airspace and then being forced to temporarily change the route due to congestion, this invention has significant advantages. By using this solution to complete the airspace capacity verification in advance before issuing the navigation path, the corresponding grid resources are locked as soon as the route is confirmed. There will be no situation where the target airspace is saturated and the aircraft cannot land. This effectively reduces the aircraft's invalid flight range, reduces energy consumption, and improves the execution efficiency and economic benefits of UAV operation navigation.

[0028] 3. This invention changes the existing interaction logic of aircraft actively transmitting full position data and autonomously calculating navigation routes. Instead, the control platform unifies the computational grid and flight paths, and issues standardized navigation commands. The onboard navigation system is only responsible for parsing and executing the path. In scenarios with massive numbers of drones flying simultaneously, this solution eliminates the need for drones to continuously upload coordinate data; they only receive brief navigation messages. The control end centrally coordinates all airspace grid resources, reducing the computational burden on the onboard navigation hardware and the data throughput burden on the platform server. In large-scale drone swarm navigation scenarios, this significantly improves the timeliness of command transmission and reception, adapting to the deployment requirements of synchronous control and navigation of thousands of drones.

[0029] 4. This invention calculates the space occupation and navigation grid size separately for different aircraft models. The navigation path meets the legal safety isolation requirements and does not require pilots to manually adjust the route. It not only meets the hard constraints of civil aviation low-altitude control regulations, but also maximizes the utilization of limited low-altitude airspace below 600 meters, achieving a balance between airspace utilization and flight compliance. Attached Figure Description

[0030] Figure 1 This is a flowchart of a navigation method based on a three-dimensional spatial gridding algorithm proposed in this invention. Detailed Implementation

[0031] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0032] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0033] Reference Figure 1This invention discloses a navigation method based on a three-dimensional spatial gridding algorithm. In this implementation case, a planned low-altitude operational airspace in a suburban area of ​​China is selected. The controlled airspace's geographical range is 116.30°-116.50° longitude and 39.80°-40.00° latitude, with a controlled altitude of 0-600m. Following the three-dimensional gridding navigation method of this invention, navigation deployment is implemented for five types of drones in three application scenarios: agricultural and forestry plant protection drones, short-haul logistics delivery drones, and power line inspection drones. The entire process includes grid division, parameter configuration, capacity calculation, route allocation, command issuance, and resource release. The implementation relies on a cloud-based airspace management platform as the main body for navigation command calculation and issuance. Various drones are equipped with standardized airborne navigation receiver terminals. These terminals do not have autonomous route planning programs; they only have command parsing and route-following flight functions.

[0034] The first step is to initialize and label the three-dimensional airspace mesh. Following the 10m side length regular hexagonal prism honeycomb three-dimensional mesh standard of this invention, the entire airspace is divided. First, the conversion rules for the X-axis (longitude), Y-axis (latitude), and Z-axis (altitude) are finalized. In this implementation case, the conversion coefficient is set as longitude and latitude values ​​multiplied by 100 to obtain the corresponding X and Y mesh numbers, and the altitude value divided by 10 and rounded down to obtain the Z mesh number. The Z value ranges from 0 to 59. The airspace is then divided into regional attributes: the 0.05° longitude and latitude range on the north side of the airspace, adjacent to the outer airspace boundary of the branch airport, is designated as the airport airspace, with a uniform mesh correction coefficient K of 0.6; the large area of ​​farmland in the middle of the airspace is a non-sensitive suburban area, with a correction coefficient K=0.8; the barren mountains and wilderness on the south side of the airspace, far from towns, airports, and important facilities, are designated as a remote uninhabited area, with K=1.0. The control platform generates unique Grid(X,Y,Z) codes for each grid based on the division results, storing all grid objects in dictionary data format. Each grid has built-in parameters including coordinate code, area type, currently occupied number of aircraft, dynamic maximum capacity, and no-fly zone sign. In this implementation case, a total of 402,000 grids were generated across the entire area. All grids were initialized to a non-no-fly zone by default, and the initial occupied capacity was reset to zero. To facilitate testing navigation performance in different scenarios, three grids near the core village locations were manually designated as permanent no-fly zones during the initial implementation phase. These no-fly zones no longer participate in any navigation route allocation; if a drone's navigation target point falls within a no-fly zone, the system directly triggers avoidance navigation logic.

[0035] The second step involves pre-entering the legally mandated basic parameters for five types of drones into the platform, establishing a mapping form between drone type, safety interval, and occupied volume. The parameters are entered strictly according to the "Interim Regulations on the Flight Management of Unmanned Aerial Vehicles": Micro-drone parameters: maximum takeoff weight 0.25kg, maximum flight speed limit 40km / h, legally mandated safety interval S=5m, calculated using the formula... Each drone occupies a safe volume of 1000 m³. Light drones have a maximum takeoff weight of 4 kg, a speed limit of 100 km / h, a span (S=10 m), and occupy a volume of 8000 m³. Small drones weigh 15 kg, have a speed limit of 120 km / h, a span (S=15 m), and occupy a volume of 27000 m³. Medium drones weigh 25 kg, have a speed limit of 150 km / h, a span (S=20 m), and occupy a volume of 64000 m³. Large drones have a takeoff weight greater than 25 kg, a speed limit of 180 km / h, a span (S=30 m), and occupy a volume of 216000 m³. The built-in calculation program on the control platform can automatically retrieve the corresponding parameters to calculate the grid capacity when it receives a navigation request from any drone model. This implementation selected five types of physical equipment—a micro aerial inspection drone, a 5 kg payload fruit and vegetable plant protection drone, a 12 kg short-distance express delivery drone, a 22 kg large forest fire prevention inspection drone, and a 30 kg large-item emergency delivery drone—to conduct navigation landing tests for each drone model.

[0036] The third step involves implementing single-grid / grid cluster capacity calculations and navigation node allocation for different drone models. First, a landing test of micro-UAV navigation is conducted: the takeoff coordinates of the micro-inspection UAV are selected as longitude 116.38°, latitude 39.88°, the target inspection point coordinates are 116.39°, 39.89°, and the target flight altitude is 45m. The platform first converts the coordinates to the corresponding grid: X=11638, Y=3988, Z=4 (40-50m altitude range). This grid is located in a suburban farmland area with K=0.8. Substituting these values ​​into the formula... =Rounded down (1×0.8)=1, the grid can accommodate a maximum of 1 micro UAV. The current grid is occupied by 0 UAVs, which meets the navigation access conditions. The platform uses this grid as the first navigation point. The subsequent target points are converted to Grid(11639,3989,4) which is free. The two independent grids are connected to form a two-point navigation route. The access navigation command is generated and sent to the airborne terminal. The command indicates the grid code to be passed, the flight speed limit of 40km / h, and the flight altitude of 40-50m. After receiving the command, the UAV smoothly completes the inspection navigation operation along the designated grid. After landing, the platform automatically reduces the number of occupied grids by 1, releasing grid resources.

[0037] Subsequently, navigation was implemented using a lightweight agricultural drone: the 5kg lightweight agricultural drone took off at coordinates 116.35°, 39.85°, with an operating altitude of 30m, corresponding to the grid Grid(11635,3985,3). The drone's S=10m, and the cluster side length is rounded up (2×10÷10)=2, requiring a total of 8 continuous grids (2×2×2) to form the navigation cluster. The platform uses the target grid as the cluster starting point and sequentially searches for 8 adjacent grids within the ranges X+1, Y+1, and Z+1. All grids are not restricted or occupied, and the entire cluster resource is then locked. The cluster as a whole serves as a single navigation node. Subsequent agricultural operation routes generate multiple continuous grid clusters according to the field operation points. Connecting the clusters end-to-end forms a complete agricultural navigation route. The navigation instructions indicate the codes of all cluster grids and a speed limit of 100km / h. The drone completes the pesticide spraying navigation flight within the three-dimensional airspace defined by the cluster. Because a single grid cannot meet the safety distance requirements for light aircraft, no instances of single aircraft flying randomly across clusters occurred throughout the entire implementation process, and the navigation boundary relied entirely on grid cluster constraints.

[0038] Small logistics drones, medium-sized forest fire prevention drones, and large emergency supply drones all use the same cluster allocation logic. For small drones (S=15m), the cluster side length is rounded up (30 / 10)=3, requiring 3×3×3=27 consecutive grid cells for a single navigation node. For medium drones (S=20m), the cluster side length is 4, requiring 64 cells per cluster. For large drones (S=30m), the cluster side length is 6, requiring 216 cells per cluster. During implementation, a congestion test was specifically set up: If one micro-drone had already accessed and occupied the same suburban grid (Grid(11640,3990,5), and a second micro-drone submitted a navigation request for the same location, the platform would calculate the grid cells. , The system determines that the grid is full, triggers the avoidance navigation logic, searches for 6 adjacent free grids above, below, left, right, front, and back of the target grid, selects Grid(11641,3990,5) as a candidate point, pushes the candidate grid coordinates in the avoidance command, and the UAV changes the navigation point to the recommended grid to complete the flight, with no flight path conflict throughout the entire process.

[0039] The fourth step is to verify the differentiated navigation between the airport airspace and remote uninhabited areas. A grid in the airport airspace north of the airport (K=0.6) is selected, and navigation for micro-drones is applied for again. Nreal = rounded down (1×0.6) = 0. No micro-drones are allowed to enter this grid for navigation, and the platform directly issues a no-entry and avoidance command to prevent aircraft from entering the airport airspace. In the remote uninhabited area to the south (K=1.0), the micro-grid Nreal = 1, and the full airspace grid navigation access conditions are more lenient, suitable for large-scale drone swarm navigation operations. In a test, 20 micro-drones sequentially applied for navigation in the same area. The platform automatically split the data into 20 adjacent free grids and assigned single-point navigation routes to each. All 20 devices operated simultaneously without airspace congestion.

[0040] Fifth, after the flight is completed, the grid resources are automatically released and the closed-loop landing is completed. Once navigation for all aircraft types ends and the aircraft leaves the corresponding grid space, the onboard terminal sends a brief departure message to the control platform. Upon receiving the message, the platform automatically decrements the corresponding grid. The previously fully loaded grid has been restored to an idle state, ready to accept new navigation requests. This implementation involved 72 consecutive hours of uninterrupted navigation testing with five types of drones rotating in rotation, executing a total of 1260 navigation routes, including 712 for micro drones, 298 for light drones, and the remaining 250 for large, medium, and small drones. No navigation route overlaps, violations of no-fly zones, or temporary detours due to airspace congestion occurred throughout the process, and all routes complied with regulatory safety separation requirements.

[0041] In addition, the implementation process is simultaneously adapted to emergency flight rerouting scenarios. When sudden temporary traffic control or temporary no-fly zones are set up in local grids, the navigation link of aircraft passing through those grids is immediately interrupted. The platform re-searches for nearby idle grid clusters in real time, generates new navigation commands, and issues them to the airborne terminal. The UAV changes its flight path according to the new commands, achieving dynamic navigation adjustment. After multi-scenario deployment verification, the navigation method of this invention, relying on the three-dimensional grid capacity dynamic calculation and passive command navigation architecture, can stably adapt to the automated navigation needs of all types of UAVs in multiple scenarios. After deployment, the safety of low-altitude airspace navigation and airspace utilization rate are improved by more than 40% compared with traditional two-dimensional autonomous navigation, fully meeting the current low-altitude flight control regulations and the standards for commercial deployment in the low-altitude economy.

[0042] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A navigation method based on a three-dimensional cubic space gridding algorithm, characterized in that, The navigation method includes the following steps: S1. Airspace Gridding: The controlled low-altitude airspace from 0 to 600m is divided into several 10m side-length regular hexagonal prism honeycomb three-dimensional grids. Each grid is uniquely identified by a Grid(X,Y,Z) three-dimensional code. The Z-axis number 0 corresponds to a height of 0-10m, and the maximum Z value of 59 corresponds to a height of 590-600m. A dynamic correction coefficient K is configured for each grid according to its geographical location attributes: K=0.6 for urban core areas / airport airspace, K=0.8 for suburbs / non-sensitive areas, and K=1.0 for remote uninhabited areas. S2, Model Parameter Pre-storage: Pre-enter the legal safety interval S corresponding to the five types of drones: micro, light, small, medium and large, and calculate the required safe space volume of a single model by using Voccupied=(2S)³; S3. Grid capacity calculation: The theoretical number of flights can be obtained by rounding down N = (single grid volume ÷ Voccupied), and the actual capacity Nreal = rounded down (N × K) can be calculated by combining the correction factor K. S4. Point-to-point grid positioning: Based on the latitude, longitude, and flight altitude of the UAV navigation starting point and target point, the corresponding three-dimensional grid code is calculated. S5. Navigation Path Allocation: Micro UAVs use individual idle grids connected in series to form navigation routes. Light, small, medium and large UAVs calculate the cluster side length according to the safety interval. A single navigation node is formed by continuous grid clusters, and multiple clusters are spliced ​​together to form a complete navigation path. S6. Passive navigation command issuance: After the control platform locks the grid resources corresponding to the route, it issues navigation commands to the UAV's onboard navigation terminal. The UAV passively receives the commands and completes navigation flight along the specified grid path. S7. Grid Resource Release: After the drone leaves the corresponding grid airspace, the control platform releases the grid resources and updates the number of times the grid has been occupied.

2. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, In step S1, the grid code conversion method is as follows: the longitude value is multiplied by the preset conversion factor to obtain the X number, the latitude value is multiplied by the preset conversion factor to obtain the Y number, and the altitude value is divided by 10 and rounded down to obtain the Z number. The control platform stores the codes, area types, no-fly status, and occupied flight data of all grids in key-value pairs.

3. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, In step S2, the legal parameters for the five types of drones are configured as follows: micro drones have a safety interval of 5m, light drones have a safety interval of 10m, small drones have a safety interval of 15m, medium drones have a safety interval of 20m, and large drones have a safety interval of 30m; micro drones have a single grid Nreal ≥ 1, while light, small, medium, and large drones have a single grid Nreal = 0, and cannot occupy a single grid to complete navigation deployment.

4. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, In step S5, the formula for calculating the side length of the mesh cluster is: =Rounded up (2S / 10), the cluster is × When assigning a cluster to a continuous cube grid, the system checks the existence and no-fly zones of all grids within the cluster sequentially. If any grid is missing or is no-fly, the cluster assignment is considered a failure.

5. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, In step S6, the navigation instructions issued by the control platform are divided into three categories: access navigation instructions, capacity reminder instructions, and avoidance and rerouting instructions. The access navigation instructions include the route grid / grid cluster code, flight altitude range, and maximum speed limit parameters for the aircraft type. The capacity indicator displays the remaining available aircraft sorties for the target grid. The evasion rerouting command is used when the target grid is fully loaded or no-fly zone, and includes recommended available alternative grid coordinates.

6. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 5, characterized in that, The method for selecting alternative grids to avoid rerouting instructions is as follows: search the target grid's six adjacent grids along the X / Y / Z axes, and select free grids that are not no-fly zones and have a current occupied number of flights less than Nreal as alternative navigation points.

7. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, It also includes temporary no-fly control steps. The control platform can set any single or multiple grids to a temporary no-fly state as needed. Grids marked as no-fly will no longer participate in the navigation route allocation for any aircraft type. If the drone's navigation target falls into a no-fly grid, the avoidance and rerouting logic will be triggered directly.

8. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, The airborne navigation terminal does not have autonomous route planning or full-domain airspace detection and calculation functions. It is only equipped with an instruction parsing module and a route following control module. The UAV executes navigation flight according to the instructions issued by the control platform throughout the entire process and does not need to transmit its own position data back to the control platform in real time.

9. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, It also includes a congestion verification step: After the drone initiates a navigation request, the platform first calculates the target grid Nreal and the currently occupied flights Nused. When Nused < Nreal, it determines that access is allowed and locks the grid resources; when Nused ≥ Nreal, it determines that the grid is full and directly outputs an avoidance and rerouting command.

10. The navigation method based on three-dimensional cubic space gridding algorithm according to claim 1, characterized in that, In step S7, after the UAV completes grid interval navigation and leaves the field, the onboard terminal sends a short departure message to the control platform. After receiving the message, the control platform automatically decrements the occupied count value of the corresponding grid by 1. The released free grid can be used for navigation resource allocation for subsequent new UAVs.

Images